Project Summary This project aims to determine how invasive closed-loop electrical stimulation can potentially rescue patterns of neural activity in the brain that give rise to successful memory functions. With an eye towards helping people affected by disorders of memory ? including the ever-growing population suffering from neurodegenerative diseases, such as Alzheimer's ? we propose to assess how functional brain dynamics can be used to guide the optimal timing, location, and properties of direct brain stimulation. Prior work has identified specific patterns of neural activity across the brain, principally spectral modulations in the theta (3-8 Hz) and gamma (30+ Hz) bands, that predict successful memory encoding and retrieval. We recently observed that the effect of stimulation on memory encoding depends on these neural signatures at the time it is delivered. Furthermore, we have shown that the functional and structural connectivity profile of a stimulation site relates to the effectiveness of stimulation on memory. Here, we aim to elucidate (1) whether state-dependent stimulation can separately be used to modulate encoding and retrieval processes, or whether common stimulation algorithms applied to both phases can more optimally restore function, (2) whether stimulation is more effective when targeted to regions with specific connectivity profiles to the medial temporal lobe, and (3) how simultaneous stimulation at multiple target sites can be optimized with knowledge of the connectivities of those sites. More broadly, this project seeks to determine whether biomarker-guided, closed-loop brain stimulation can be used to correct pathologic brain states and restore the function of an impaired memory system.